Presentation is loading. Please wait.

Presentation is loading. Please wait.

BY: ALI AJORIAN ISFAHAN UNIVERSITY OF TECHNOLOGY 2012 GPU Architecture 1.

Similar presentations


Presentation on theme: "BY: ALI AJORIAN ISFAHAN UNIVERSITY OF TECHNOLOGY 2012 GPU Architecture 1."— Presentation transcript:

1 BY: ALI AJORIAN ISFAHAN UNIVERSITY OF TECHNOLOGY 2012 GPU Architecture 1

2 Age of parallelism Single CPU performance  Doubled every 2 years for 30 years until 5 years ago.  Marginal improvement in the last 5 years. 2005 year and checking walls  Memory Wall  Power Wall  Processor Design Complexity Sequential or parallel: this is the problem!!!  More cores rather than more clock rate 2

3 Early parallel computing It was not a big idea  Main frames and super computers 3

4 And now GPUs Stands for “Graphics Processing Unit” Integration Scheme: a card on the motherboard with Massively Parallel computing power 4

5 A desktop supper computer 5

6 History of parallel computing 6

7 GPUs: A Brief History Stage0: graphic accelerators  Early VGA cards accelerate 2D GUI  Just configurable Stage1: Fixed Graphics Hardware  Graphics-only platform  Very limited programmability Stage2: GPGPU  Trick GPU to do general purpose computing  Programmable, but requires knowledge on computer  graphics Stream Processing Platforms  High-level programming interface  No knowledge on Computer Graphics is required  Examples: NVIDIA’s CUDA, OpenCL 7

8 Stream Processing Characteristics Fairly simple computation on huge amount of data (streams)  Single Program Multiple Data (SPMD) Data Parallelism  e.g., Matrix Operations, Image Processing 8

9 Graphic accelerators to CUDA GPUs(cont) 9

10 CUDA programming model CPU + GPU heterogeneous programming  Applications with sequential and parallel parts Host : CPU  Sequential threads Device : GPU  Parallel threads in SIMT architecture  some kernels that runs on a grid of threads. 10

11 CUDA programming model 11

12 CUDA programming model(cont) 12

13 GPU Architecture (NVIDIA) 13

14 GPU Architecture (Fermi) 14

15 SM architecture 15

16 CUDA programming model 16

17 Memory types Per block  registers  shared memory Per thread  local memory Per grid  Global memory  Constant memory  Texture memory 17

18 Memory types(cont) 18

19 Questions? 19


Download ppt "BY: ALI AJORIAN ISFAHAN UNIVERSITY OF TECHNOLOGY 2012 GPU Architecture 1."

Similar presentations


Ads by Google